3 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
4 |
============================== |
============================== |
5 |
|
|
6 |
|
checkpoint51f_pre |
7 |
|
o Added on-the-fly spatial interpolation capability |
8 |
|
"USE_EXF_INTERPOLATION" to pkg/exf. |
9 |
|
- This is a temporary Cartesian-grid hack until |
10 |
|
the super-duper ESMF coupler becomes available. |
11 |
|
- See verification/global_with_exf/README for usage example. |
12 |
|
- Removed obsolete EXFwindOnBgrid and SEAICEwindOnCgrid |
13 |
|
flags and modified pkg/seaice accordingly. |
14 |
|
o Bug fix to pkg/ptracers, pkg/generic_advdiff/gad_calc_rhs.F, |
15 |
|
and pkg/kpp/kpp_transport_ptr.F for dealing with tracer |
16 |
|
non-local transport term. |
17 |
|
|
18 |
|
checkpoint51e_post |
19 |
|
o pkg/mom_vecinv: |
20 |
|
- changes related to hFac: use S/R from aim.5l_cs/code (coriolis & vort.advec) |
21 |
|
- add time-ave diagnostic of vorticity advection. |
22 |
|
o add diagnostic (instantaneous) of Shapiro Filter effect for T,S & UV. |
23 |
|
o atmospheric geopotential: - include water vapor effect; |
24 |
|
- use p^kappa in p* scaling. |
25 |
|
o land pkg: read initial state from files. |
26 |
|
o AIM : add stratospheric drag in the upper level |
27 |
|
|
28 |
|
checkpoint51d_post |
29 |
|
o change function ILNBLNK (=> last non-blank char.) to return zero |
30 |
|
(instead of the string length) when the string. char is empty |
31 |
|
o fix bug in dxV,dyU definition of cubed-sphere grid (ini_curvilinear_grid) |
32 |
|
o modif's to restore TAMC compatibility |
33 |
|
- avoid "_d" constructs in PARAMETER declarations |
34 |
|
- avoid interrupting lines by comment lines |
35 |
|
o Use cluster local disks for purely local I/O |
36 |
|
vs. globally visible disks needed for ctrl stuff |
37 |
|
(merged from ecco-branch) |
38 |
|
- new active_file_loc... and mdsio_..._loc routines |
39 |
|
- affected packages: autodiff, ctrl, grdchk, mdsio |
40 |
|
|
41 |
|
checkpoint51c_post |
42 |
|
o introducing integer flag debugLevel |
43 |
|
- debLevZero: suppress all output |
44 |
|
- debLevA: standard |
45 |
|
- debLevB: equivalent to current debugMode=.TRUE. |
46 |
|
o introducing pathname variable mdsioLocalDir for mdsio |
47 |
|
that can be specified at runtime |
48 |
|
(relevant for cluster I/O to local disk) |
49 |
|
` |
50 |
|
>>>checkpoint51b_post |
51 |
|
adjoint of ptracers |
52 |
|
o disentangled ALLOW_PTRACERS using new ALLOW_GCHEM |
53 |
|
o extended state in checkpoint_lev?, etc. |
54 |
|
o alternative ctrl/cost with tr1 or ptracer |
55 |
|
o new comlev1_bibj_ptracers |
56 |
|
o new ptracers.flow |
57 |
|
o made ptracers_forcing same as external_forcing_s |
58 |
|
o added verification for adjoint with ptracers |
59 |
|
- adm. grdchk fail so far |
60 |
|
- tlm. grdchk OK |
61 |
|
- checked tapelev/comlev recomputations: OK |
62 |
|
<<<checkpoint51b_pre |
63 |
|
|
64 |
|
checkpoint51a_post |
65 |
|
o added code for biogeochemistry package |
66 |
|
- new pkgs: dic, gchem |
67 |
|
- new subroutine in ptracers |
68 |
|
o modify code for biogeo packages |
69 |
|
- in model/src and pkg/ptracers |
70 |
|
|
71 |
|
checkpoint51 |
72 |
|
o merged improved packages from ecco-branch (ecco_c51_e34) to MAIN |
73 |
|
- autodiff, ctrl, exf, grdchk, seaice |
74 |
|
kept separate packages |
75 |
|
- cost, ecco, obcs |
76 |
|
o updating verification/internal_wave, verification/lab_sea |
77 |
|
(all verif's pass on faulks.lcs.mit.edu RH7.3, g77) |
78 |
|
|
79 |
|
checkpoint50i_post |
80 |
|
o Preparing next differentiable checkpoint and sync |
81 |
|
of MAIN vs. ecco-branch |
82 |
|
- 1st step: this one is differentiable |
83 |
|
- 2nd step: next(! not this) checkpoint will have updated packages |
84 |
|
from ecco-branch) |
85 |
|
|
86 |
|
checkpoint50h_post |
87 |
|
o new package "land": provide surface BC for AIM. |
88 |
|
o pkg/ptracers: apply exchange to all tracers after reading a pickup. |
89 |
|
o AIM: use true log-P extrapolation for near surface temp (= as in |
90 |
|
F.Molteni paper) ; update aim.5l_cs output. |
91 |
|
o New variable in PARM03 for more flexible pickups |
92 |
|
o somewhat cleaned package initialisation sequence for |
93 |
|
ctrl/ cost/ ecco/ in packages_... routines |
94 |
|
|
95 |
|
checkpoint50g_post |
96 |
|
o merged with release1_p17 (pkg/seaice and verification/lab_sea) |
97 |
|
- added SEAICE_MULTILEVEL for 8-category sea-ice thermodynamics |
98 |
|
- LSR sea-ice dynamic solver moved to SouthWest B-grid location and |
99 |
|
made the default because of faster convergence than ADI |
100 |
|
|
101 |
|
checkpoint50f_post |
102 |
|
o exf totally restructured |
103 |
|
- replaced all exf_set_ routines by one generic exf_set_gen.F |
104 |
|
and made sure it's still differentiable |
105 |
|
- split exf_getffields.F into new exf_bulkformulae.F |
106 |
|
- exf_inscal_/exf_outscal_ suffixes consistent with field names |
107 |
|
- updated adjoint/makefile and relevant verif. accordingly |
108 |
|
|
109 |
|
checkpoint50f_pre |
110 |
|
o AIM: near surface temp evaluated from Pot.Temp: fixed for Part Cell ; |
111 |
|
+ change Evap over land. (update aim.5l_cs output) |
112 |
|
o added missing .cvsignores and made uniform across expts |
113 |
|
o changed configuration of tidal_basin_2d expt |
114 |
|
o added pkg/debug functions for tracing code in fatal crashes |
115 |
|
o tidied pkg/monitor to use MONITOR_OPTIONS.h |
116 |
|
o new fn in pkg/monitor to stop model if solution grows too big |
117 |
|
o removed lots of single quotes from comments |
118 |
|
|
119 |
|
checkpoint50e_post |
120 |
|
o make KPP work with PTRACERS |
121 |
|
- fix gad_calc_rhs to call new routine kpp_transport_ptr, which is |
122 |
|
nearly a copy of kpp_transport_s |
123 |
|
- there is no analogue to SurfaceTendencyS, so I have to use |
124 |
|
gPtr(of the surface layer) instead |
125 |
|
o add a new platform SunFire+mpi (SunFire 15000) to genmake |
126 |
|
checkpoint50e_pre |
127 |
|
|
128 |
|
checkpoint50d_post |
129 |
|
o change kpp output from multiple-record state files to single-record state |
130 |
|
files analogous to write_state.F |
131 |
|
o reduce the output frequency of cg3d-related stuff to the monitor frequency, |
132 |
|
analogous to the cg2d-related output. |
133 |
|
o fix small problem with in ptracers_write_checkpoint.F: len(suff)=512, |
134 |
|
so that writing to internal file fn (with length 512) fails. |
135 |
|
checkpoint50d_pre |
136 |
|
|
137 |
|
checkpoint50c_post |
138 |
|
o fix problem in ini_procs.F when using MPI (was there since checkpoint48f) |
139 |
|
o add monitor diagnostic of vorticity (mon_vort3.F) and potential energy. |
140 |
|
o add time-average diagnostic of U*V and vertical diffusive flux for theta. |
141 |
|
o bug fixed in convective adjustment diagnostic (multi-tile set-up). |
142 |
|
o 4th Order Adv scheme: return to previous (c40pre7) masking near boundaries. |
143 |
|
|
144 |
|
checkpoint50c_pre |
145 |
|
Merging from release1_p13: |
146 |
|
o bug fix for pkg/seaice dynamic solver |
147 |
|
o Added SEAICE_initialHEFF to pkg/seaice |
148 |
|
|
149 |
|
checkpoint50b_post |
150 |
|
o store u*,v* in gU,V instead of in guNm1,gvNm1 |
151 |
|
o in order to put the momForcing out of the Adams-Bashforth: |
152 |
|
move forcing & CD-scheme calls from mom_fluxform & mom_vecinv |
153 |
|
to timestep.F |
154 |
|
o new flag "useCDscheme" (default=F); replace guCD,gvCD by local arrays |
155 |
|
o re-generate output for highly sensitive test-exp: |
156 |
|
global_ocean_pressure, ideal_2D_oce, lab_sea, natl_box |
157 |
|
|
158 |
|
checkpoint50b_pre |
159 |
|
o restore the use of momAdvection & useCoriolis flags (mom_vecinv) |
160 |
|
o corrected CPP option mismatch in exf_Bulk... functions for case |
161 |
|
#define ALLOW_ATM_WIND && #undef ALLOW_ATM_TEMP |
162 |
|
|
163 |
|
checkpoint50a_post |
164 |
|
o Bug fix for merging between c50 and KPP. |
165 |
|
ikey was passed from thermodynamics to kpp_calc via |
166 |
|
common block rather than being recomputed in kpp_calc, |
167 |
|
in contradiction with new key itdkey. |
168 |
|
New key ikppkey created, and tamc.h headers updated. |
169 |
|
|
170 |
|
checkpoint50 |
171 |
|
- 2nd step of merge with ecco-branch: |
172 |
|
merge changes in ecco-branch back into main trunk |
173 |
|
- bug fix in find_alpha |
174 |
|
- more storing in gmredi_calc_tensor.F, kpp_routines.F |
175 |
|
- avoid iterative loop in ini_pressure for ALLOW_AUTODIFF_TAMC |
176 |
|
- merged some sliced AD support routines |
177 |
|
|
178 |
|
checkpoint49 |
179 |
|
merged from ecco-branch: |
180 |
|
o exf: |
181 |
|
- Enable initialisation of forcing fields to constant |
182 |
|
(runtime) values. |
183 |
|
- in exf_getffields.F |
184 |
|
Reduce i-/j-loop to interior domain, discarding overlaps. |
185 |
|
That also fixes wrong TAF-key computations for key_1, key_2 |
186 |
|
with bulf formulae. |
187 |
|
- exf_init.F modify #ifdef for exf_init_evap |
188 |
|
- exf_getffieldrec.F, ctrl_getrec.F |
189 |
|
The following INT-usages are not safe: |
190 |
|
fldsecs = int(fldsecs/fldperiod)*fldperiod |
191 |
|
fldcount = int(fldsecs/fldperiod) + 1 |
192 |
|
and were modified. |
193 |
|
o autodiff: checkpoint_lev... |
194 |
|
- updated storing for bulk formulae (some arrays were missing) |
195 |
|
- added state for seaice package |
196 |
|
- synchronized with main branch |
197 |
|
o DIVA: |
198 |
|
Modif.s in |
199 |
|
- the_main_loop.F |
200 |
|
- the_model_main.F |
201 |
|
and added cost_final_restore.F cost_final_store.F |
202 |
|
o With these changes, it is possible to fully separate |
203 |
|
the first forward run to compute the cost function (mdthe_main_loop) |
204 |
|
and adjoint runs over individual DIVA intervals (adthe_main_loop) |
205 |
|
into separate model executions. |
206 |
|
This enables better tuning of DIVA to given queue sizes. |
207 |
|
o Loss of cost function value in consecutive DIVA runs is |
208 |
|
fixed by storing to 'costfinal' file. |
209 |
|
o Appropriate call of ctrl_unpack/_pack,grdchk_main at |
210 |
|
beginning/end of adjoint integration is accounted for. |
211 |
|
The current changes apply to TAF-1.4.26. |
212 |
|
TAF has one remaining bug, requiring one manual change |
213 |
|
(see MITgcm manual). |
214 |
|
o cal: synchronized between MAIN and ecco-branch |
215 |
|
|
216 |
|
checkpoint48i_post |
217 |
|
o Changes to restore differentiability of code w.r.t. previous tag |
218 |
|
(mostly adding new routines to make list and replacing |
219 |
|
pressure by totPhiHyd). |
220 |
|
o Updated and merged grdchk package |
221 |
|
- has both ADM and TLM checks |
222 |
|
- works for single- and multi-proc. |
223 |
|
- output cleaned |
224 |
|
- worked successfully for parallel DIVA |
225 |
|
o genmake: |
226 |
|
- Adding -platform=SP4 |
227 |
|
- Modif. for DEC Alpha (many don't have makedepend; use mkdep instead). |
228 |
|
o Modif. for DIVA: |
229 |
|
ctrl_unpack should not be called in intermediate DIVA interval. |
230 |
|
(similar thing would apply for active files such as |
231 |
|
tbar, sbar, psbar, which should not be reinitialised; |
232 |
|
only relevant for ecco-branch) |
233 |
|
|
234 |
|
checkpoint48h_post |
235 |
|
o improve "ini_p_ground" when using finite difference form for calc_phi_hyd |
236 |
|
o Added net flux to downward flux conversion to pkg/exf/exf_getffields.F |
237 |
|
o Added SEAICE_initialHEFF to pkg/seaice |
238 |
|
o ecco_the_model_main.F: SEAICE_WRITE_PICKUP must precede WRITE_CHECKPOINT |
239 |
|
|
240 |
|
checkpoint48g_post |
241 |
|
--- EOS: |
242 |
|
o compute locally the pressure for use in EOS: UNESCO, JMD95P, MDJWF |
243 |
|
store total Potential in totPhyHyd for diagnostics & EOS. |
244 |
|
o fix restart and overlap Pb when using Z-coord and EOS funct. of P |
245 |
|
--- Pressure/geopotential gradient: |
246 |
|
o use Ro_surf & R_low instead of hFac in cal_phi_hyd: |
247 |
|
o NLFS + p-coord.: eta*Alpha' is not longer lagging 1.time-step behind |
248 |
|
( changes results of global_ocean_pressure ) |
249 |
|
o NLFS + z-coord.: add eta*Rho' contribution. |
250 |
|
o accurate phiHyd when using interface-W at the middle between 2 centers |
251 |
|
( changes results of ideal_2D_oce ) |
252 |
|
o includes r* 2nd term in potential gradient. |
253 |
|
|
254 |
|
checkpoint48f_post |
255 |
|
Merging from release1_p12: |
256 |
|
o Modifications for using pkg/exf with pkg/seaice |
257 |
|
- improved description of the various forcing configurations |
258 |
|
- added basic radiation bulk formulae to pkg/exf |
259 |
|
- units/sign fix for evap computation in exf_getffields.F |
260 |
|
- updated verification/global_with_exf/results/output.txt |
261 |
|
o Added pkg/sbo for computing IERS Special Bureau for the Oceans |
262 |
|
(SBO) core products, including oceanic mass, center-of-mass, |
263 |
|
angular, and bottom pressure (see pkg/sbo/README.sbo). |
264 |
|
o Lower bound for viscosity/diffusivity in pkg/kpp/kpp_routines.F |
265 |
|
to avoid negative values in shallow regions. |
266 |
|
- updated verification/natl_box/results/output.txt |
267 |
|
- updated verification/lab_sea/results/output.txt |
268 |
|
o MPI gather, scatter: eesupp/src/gather_2d.F and scatter_2d.F |
269 |
|
o Added useSingleCpuIO option (see PARAMS.h). |
270 |
|
o Updated useSingleCpuIO option in mdsio_writefield.F to |
271 |
|
work with multi-field files, e.g., for single-file pickup. |
272 |
|
o pkg/seaice: |
273 |
|
- bug fix in growth.F: QNET for no shortwave case |
274 |
|
- added HeffFile for specifying initial sea-ice thickness |
275 |
|
- changed SEAICE_EXTERNAL_FLUXES wind stress implementation |
276 |
|
o Added missing /* */ to CPP comments in pkg/seaice, pkg/exf, |
277 |
|
kpp_transport_t.F, forward_step.F, and the_main_loop.F |
278 |
|
o pkg/seaice: |
279 |
|
- adjoint-friendly modifications |
280 |
|
- added a SEAICE_WRITE_PICKUP at end of the_model_main.F |
281 |
|
|
282 |
|
checkpoint48e_post |
283 |
|
o New S/R for diagnostic of bottom pressure (phi0surf was missing in c48d_post) |
284 |
|
o dynamics: change definition of computational domain & adapt mom_fluxform |
285 |
|
accordingly ; when viscA4=0, allows to run the dynamics with Olx=Oly=2. |
286 |
|
o following exp works with Olx=Oly=2 (and pass through checkbounds option): |
287 |
|
adjust_nlfs.cs-32x32x1 adjustment.128x64x1 adjustment.cs-32x32x1 |
288 |
|
exp0 exp1 exp2 exp4 exp5 hs94.cs-32x32x5 hs94.128x64x5 hs94.1x64x5 |
289 |
|
global_ocean.90x40x15 global_with_exf global_ocean_pressure |
290 |
|
inverted_barometer solid-body.cs-32x32x1 |
291 |
|
(high order AdvScheme and/or multiDimAd works with Olx=Oly=2 ? not sure) |
292 |
|
|
293 |
|
checkpoint48d_post |
294 |
|
o in preparation for r*: |
295 |
|
a) use pre-computed gradient of hydrostatic potential: |
296 |
|
changes in timestep.F & mom_cdscheme.F affects results of ideal_2D_oce |
297 |
|
b) move phi0surf from calc_phi_hyd to calc_grad_phi_hyd: |
298 |
|
=> affects results of glob_oce_pressure (different truncation error) |
299 |
|
update output.txt of ideal_2D_oce & global_ocean_pressure |
300 |
|
|
301 |
|
checkpoint48d_pre |
302 |
|
o in preparation for r*: |
303 |
|
new S/R (calc_grad_phi_hyd.F) to compute Hydrostatic potential gradient. |
304 |
|
pass the 2 comp. of the grad. as arguments to momentum S/R. |
305 |
|
but for the moment, only used if it does not change the results. |
306 |
|
o add finite volume form of calc_phi_hyd for buoyancyRelation OCEANIC & OCEANICP |
307 |
|
|
308 |
|
checkpoint48c_post |
309 |
|
Modifications in treatment of initial flux forcing controls |
310 |
|
to avoid extra storing of flux fields. |
311 |
|
|
312 |
|
checkpoint48c_pre |
313 |
|
o Added tidal_basin_2d experiment (currently 1D!) |
314 |
|
o can use r* Coordinate with OBC (but Orlanski Rad. BC not yet implemented) |
315 |
|
|
316 |
|
checkpoint48b_post |
317 |
|
o r* coordinate added in #ifdef NONLIN_FRSURF block. |
318 |
|
(modification to pressure gradient not yet implemented) |
319 |
|
o mom_fluxform: new S/R mom_calc_rtrans.F computes vertical transports |
320 |
|
(for U & V) that are used for vertical advection of momentum. |
321 |
|
|
322 |
|
checkpoint48a_post |
323 |
|
o In preparation for r*, use temp. array to compute vertical transport |
324 |
|
in flux-form momentum pkg. => change results of few test-exp. |
325 |
|
Update output of: exp4, exp5, front_relax, global_ocean_pressure, |
326 |
|
lab_sea, natl_box & plume_on_slope |
327 |
|
|
328 |
|
checkpoint48 |
329 |
|
o fix small bug that was causing lot of warning (TIMER_STOP 'UPDATE_CG2D') |
330 |
|
|
331 |
|
checkpoint47j_post |
332 |
|
o add new exp. (ideal_2D_oce) to test "exotic" parameters and options. |
333 |
|
o moved adexch from forward_step to addummy_in_stepping |
334 |
|
o added exch in ctrl_map_... |
335 |
|
o aveFreq.=0. at end of cost_final to avoid overwrite in |
336 |
|
reverse checkpointing loop |
337 |
|
o Yet more changes in gmredi: |
338 |
|
* adgmredi_calc_tensor |
339 |
|
avoiding all recomputation of gmredi_slope_limit |
340 |
|
* adgmredi_x/y/rtransport |
341 |
|
added flag for excessive storing to avoid recomp. of |
342 |
|
u/v/rtans, dTdx/y/z |
343 |
|
-> this is not really necessary and very memory-consuming |
344 |
|
* adgmredi_slope_psi: |
345 |
|
consistency with gmredi_slope_limit in treatment of GM_slopeSqCutoff |
346 |
|
* gmredi_slope_limit |
347 |
|
re-activated full calculation of taperfct for case 'ac02' |
348 |
|
o updating verification/ adjoints of carbon, global_ocean.90x40x15 |
349 |
|
|
350 |
|
checkpoint47i_post |
351 |
|
o GM_EXCLUDE_() replace GM_TAPER_(): by default, all the code is compiled. |
352 |
|
o add GM_Small_Number & GM_slopeSqCutoff as run-time parameters (data.gmredi) |
353 |
|
default values = 1.D-12 & 1.D+48 (=> give same results as checkpoint47f_post) |
354 |
|
o cg2dTargetResWunit: define tolerance using deltaTmom (and not deltaTfreesurf) |
355 |
|
o gmredi: fix few bugs. |
356 |
|
|
357 |
|
checkpoint47h_post (duplicate tag for PH's c48 tag, in order to move c48) |
358 |
|
This will be a good checkpoint to plug into the ecco-branch. |
359 |
|
o few modif.'s to get latest version adjointed |
360 |
|
(mainly kick out code in ini_linear_phisurf) |
361 |
|
o modif's to run adjoint with exactConserv |
362 |
|
o case GM_BOLUS_ADVEC should be cleaned |
363 |
|
S/R gmredi_slope_psi should be cleaned |
364 |
|
o verification/carbon now has exactConserv=.TRUE. |
365 |
|
|
366 |
checkpoint47g_post |
checkpoint47g_post |
367 |
o time-average AIM output: becomes independent from statvars time-average |
o time-average AIM output: becomes independent from statvars time-average |
368 |
(can be used with #undef ALLOW_TIMEAVE) ; add new variables (e.g., RH) |
(can be used with #undef ALLOW_TIMEAVE) ; add new variables (e.g., RH) |